DE10351957A1 - Damping system with a LOLIMOT model against drive train vibrations - Google Patents

Abstract

The invention relates to a damping device having a detection device (4, 7) for determining a mechanical state variable (DELTAalpha¶MODELL¶, DELTAalpha¶IST¶) representing the torsion of a drive train (3) of an internal combustion engine (1) and an actuating device (2) for driving an internal combustion engine (1) with a manipulated variable as a function of the mechanical state variable (DELTAalpha¶MODELL¶, DELTAalpha¶IST¶). It is proposed that the mechanical quantity of state (DELTAalpha¶MODELL¶, DELTAalpha¶IST¶) be determined by a predictor member (4) containing a model of the drive train (3) and / or the internal combustion engine (1).

Description

The The invention relates to a damping device according to the generic term of claim 1 and a damping method according to the generic term of claim 12.

By technical improvements, especially in direct injection technology could the dynamics of the power delivery of internal combustion engines be increased significantly. This leads to pronounced load jumps in powertrains of motor vehicles that use these internal combustion engines for propulsion. load jumps provide a wide excitation in the frequency domain for the oscillatory system Driveline dar. This can Low-frequency torsional vibrations in the drive train are triggered. The eigenform of the deepest torsional vibration consists of an angular rotation of the motor relative to the driven wheels. A such oscillation turns out to be particularly jerky in the longitudinal direction of the Vehicle noticeable and reduces the drivability of the motor vehicle considerably. Furthermore, these vibrations as well as the load jumps themselves a heavy burden for the drive train, whereby the wear is increased and it comes to material fatigue can.

A known possibility to suppress the vibrations and their negative effects exists in it, the vibration from one of a speed sensor on the Internal combustion engine recorded measuring signal filter, and by the internal combustion engine to apply a counter torque to the vibration. For this purpose, the signal of the speed sensor is filtered with a low-pass filter and out of phase.

The However, the method described has the disadvantage that it is close operated the stability limit must be in order to be effective. In particular, the problem here is that the damping torque is applied at a frequency that is the torsional resonance frequency equivalent. Because of that lead already small errors in the calculation of the counter torque or small changes in the mechanical behavior of the powertrain under certain circumstances Instabilities. It is important to take into account that the mechanical properties of the powertrain in the Generally about change the life of a motor vehicle, for example, comes it to wear on gears or to a change the elastic properties of shaft couplings. Another disadvantage of the method is that only on existing vibrations can be reacted, the damping Only sets in when the high load on the powertrain already is available.

Of the The invention is therefore based on the object, with the least possible effort vibrations in Suppress drive train, in particular, high loads on the drive train and jerking movements of the vehicle should be avoided.

The Task is with a damping device according to claim 1 and a damping method according to claim 12 solved.

The Invention is based on the physical knowledge that the internal combustion engine, the drive train or the speed sensor have a dead time, which is the regulation of damping torques for suppression of torsional vibrations in the drive train difficult. For example leads one increased Fuel supply is not directly related to increased drive torque Internal combustion engine, since the fuel quantity injected clocked into the combustion chambers becomes, which causes time losses.

advantageously, Therefore, in the context of the invention, a predictor member is used to a mechanical state variable of the drive train as Determine response to a manipulated variable. This has the advantage that the manipulated variable as a function of the determined fixed mechanical state variable can be and the internal combustion engine with the so modified Control value activated becomes. Thus, the excitation of torsional vibrations is already suppressed.

The Manipulated variable for the internal combustion engine can For example, be the fuel supplied to the engine. However, it is also conceivable other manipulated variables, such as the Affect throttle position.

The mechanical state variable gives preferably the temporal change the torsion of the drive train again, to torsional vibrations clearly from the others usual in the company Distinguish loads.

The inventive device considered preferably the set transmission ratio of Gear and other translations in the Powertrain. Thus, the damping device a signal input for receiving a the transmission ratio of Gear reproducing signal include.

The predictive member preferably has a model of the internal combustion engine and the drive stranded to determine the mechanical state quantity. A model has the advantage that it allows a mathematical prediction of the mechanical response to given controls.

Preferably is the one contained in the predictor Model essentially dead time free. Especially the internal combustion engine Due to the combustion process has a dead time, this has the Advantage of a time saving. Is before a control intervention the actual Response of the powertrain to the manipulated variable, so can during the while passing dead time further vibration stimulating impulses given by the manipulated variable without being regulated. Will be the answer timely, i. as fast as the arithmetic unit of the model allows, calculated, so can Torsionsschwingungen be suppressed in the initial stage or the excitation of torsional vibrations can be suppressed. Also, the powertrain or the speed sensor can a Have dead time.

advantageously, indicates the model of the predictor member adaptable model parameters. This has the advantage that the Model can be adjusted if the mechanical properties the internal combustion engine, powertrain or vehicle. For example can the drive train due to wear its mechanical Properties over change the lifting time of the motor vehicle or by loading the Motor vehicle, the mass of the motor vehicle can be changed. Preferably, the damping device Therefore, an adaptation unit for adapting the model parameters while driving on. This allows the model at any time changed mechanical properties be adjusted. It is also conceivable to change the parameters as a function of the speed, for example by one higher Air resistance to be considered. In an advantageous embodiment of the invention the adaptation unit the model states. The adaptation unit can to measured values, which are measured anyway in the vehicle, such as the speed of the internal combustion engine or the driven wheels use. This can disturbances and Model inaccuracies are corrected immediately, which improves the quality of the prediction of predictor elevated.

In an advantageous embodiment The invention provides a multi-mass model of the engine and the drive train with the vehicle mass in the model dar can depending on the required Modeling accuracy two, three, four or more model masses by spring / damper elements be connected to each other. This has the advantage that the oscillatory system Although simplified from internal combustion engine, powertrain and vehicle mass but still close to reality can be displayed.

Around to detect the first torsional vibration form of the drive train may advantageously be a two-mass torsional vibration model be used as a multi-mass model. The two-mass torsional vibration model includes two mass moments of inertia and a spring / damper element, these moments of inertia combines. The spring / damper element consists of a torsion spring and a torsion damper. However, it is also conceivable only to arrange a torsion spring and the damping of the drive train unconsidered which would simplify the model. This is especially true at very small attenuations of the drive train advantageous. Preferably, the first moment of inertia represents the moving parts of the internal combustion engine, for example, the Crankshaft, the connecting rods and the pistons. With the second moment of inertia become the wheels and the vehicle mass modeled, wherein the vehicle mass in the Calculation of the moment of inertia with a radius of gyration enters, which is substantially equal to the radius of the wheels. The model considered in each case the set transmission ratio of Transmission. Alternatively, it is also conceivable, in addition further moments of inertia for example to introduce the gearbox into the model, resulting in the prediction accuracy of the model can.

Advantageously, the model parameters are constants of the mass moment of inertia or of the spring / damper element. Thus, one model parameter may represent the value of one moment of inertia, the other model parameter the value of the other mass moment of inertia, a third model parameter the torsion spring constant of the drive train and a fourth model parameter the torsional damping constant of the drive train. In conjunction with the adaptation unit for adapting the model parameters, the two-mass torsional vibration model can thus be adapted to changed mechanical properties of the drive train and other components. Thus, for example, the moment of inertia that models the wheels with the vehicle mass can be increased by a corresponding amount if there is a higher payload of the vehicle. This has the advantage that the prediction accuracy of the model does not suffer from a change in vehicle mass. Furthermore, it is conceivable that the spring stiffness of the drive train changes over the life of the vehicle. In such a case, the torsion spring constant of the drive train can advantageously be adapted to the changed mechanical properties via the adaptation unit. In addition, the torsion damper can also constant adapted to the adaptation unit changed mechanical conditions. In a further advantageous embodiment, the model parameters are determined as a function of the set transmission ratio of the drive train. Advantageously, a set of model parameters is maintained in a memory for each gear ratio. When changing the gear ratio, the possibly adapted model parameters of the last gear ratio are stored again. This has the advantage that the predictive member has a suitable model immediately when specifying a changed gear ratio.

In a further preferred embodiment is the model that is in the predictor is included, a LOLIMOT model (Local Linear Model Tree), which the mechanical behavior of the internal combustion engine or the drive train with vehicle mass on the basis of predetermined linear imaging functions represents. As input variables for the LOLIMOT model For example, come the Torsionsdrehgeschwindigkeiten the internal combustion engine and the wheels and the predetermined torque of the internal combustion engine into consideration. Alternatively, it is also conceivable to take into account other variables, for example the fuel supply to the engine or the throttle position. The LOLIMOT model preferably calculates a predicted one from the input variables Difference of the rotational angular velocities of the internal combustion engine and the wheels or another variable that reflects the torsion of the powertrain. The LOLIMOT model linked doing several linear mapping functions, the partially approximately linear behavior the internal combustion engine and the drive train play.

Advantageously, the LOLIMOT model has weighting functions associated with the individual mapping functions. The weighting functions may be, for example, Gaussian normal distributions with which the linear mapping functions are multiplied. LOLIMOT models with linear mapping functions and weighting functions are closer in the publication "Local Linear Model Trees (LOLIMOT) Toolbox for Nonlinear System Identification" of 0. Nelles et al. (12 ^th IFAC Symposium on System Identification, St. Barbara, United States, 2000 The combination of linear mapping functions and weighting functions allows on the one hand a slight adaptation of the model to the actual mechanical properties and also offers advantages in terms of computing speed, since no computation-intensive non-linear equations have to be calculated ,

Preferably The weighting function has parameters that act as model parameters for example, can be adapted by the adaptation unit. In order to can be the weighting of individual linear mapping functions be adjusted. Another advantage is that the selectivity of the single weighting functions can be changed, thereby in certain circumstances a better adaptation of the model to reality is possible.

advantageously, become the parameters of the linear mapping functions of the LOLIMOT model adapted as model parameter by the adaptation unit. This can for example about a regression algorithm, as described in the above-mentioned publication from 0. Nelles et al. is described. Furthermore, it is conceivable, more Imaging functions and weighting functions in the context of adaptation to generate the model to the mechanical behavior of the powertrain better adapt.

The The invention further comprises a motor controller with a damping device in one of the described embodiments. Such an engine control is particularly suitable for the internal combustion engine to control so that wear-increasing load peaks and jerking movements in the longitudinal direction of the vehicle can be avoided.

Of Furthermore, the invention comprises a damping method which, for example with one of the described damping devices carried out can be.

The Invention will be described below with reference to the accompanying drawings. Show it:

1 a schematic representation of a damping device according to the invention,

2 a schematic representation of a LOLIMOT model and

3 a schematic representation of a two-mass torsional vibration model.

1 schematically shows a control circuit equivalent circuit diagram, in which an internal combustion engine 1 from an actuator 2 is controlled. In the drawing, it is shown that the manipulated variable with which the internal combustion engine 1 from the actuator 2 is controlled, the amount of fuel m is an injection process. Actually, the adjusting device 2 further parameters of the internal combustion engine 1 control, for example, the throttle position.

The internal combustion engine 1 drives over a powertrain 3 the wheels of a vehicle. The powertrain 3 includes multiple shafts, a gear, a differential and joints for torque transmission between the individual components. The powertrain 3 is from the internal combustion engine 1 with the moment M _being driven.

The adjusting device 2 represents the amount of fuel to be injected m according to the specification of the drive torque M ' _SOLL the internal combustion engine 1 one. The adjusting device 2 uses a control method that is well known to the skilled person in various embodiments.

The damping device comprises a predictor member 4 , which is a model of the internal combustion engine 1 and the powertrain 3 contains. The model is a torsional oscillator with two mass moments of inertia and a torsion spring damper between the two mass moments of inertia. In this case, a moment of inertia corresponds to the mass moment of inertia of the moving parts of the internal combustion engine 1 , The torsion spring element provides the drive train 3 with its components. The second mass moment of inertia of the model corresponds to the driven wheels and the mass of the vehicle, which enter into the calculation of the second mass moment of inertia with a radius of gyration corresponding to the radius of the wheels. M ' _SOLL is applied to the model as a load moment. The predictor member 4 calculates therefrom on the basis of the model, the angular velocity of the shaft of the internal combustion engine 1 at the powertrain 3 is connected, and the angular velocity of the driven wheels. Here, the model takes into account the set transmission ratio of the transmission. The output of the predictor member 4 contains a signal which represents the difference Δα _{MODEL of} the described angular velocities.

The difference Δα _MODEL corresponds to the temporal change of the torsion of the drive train 3 between the internal combustion engine 1 and the driven wheels. In order to suppress a torsional vibration as effectively as possible, according to a classical mechanical damping of a PD member 5 a damping _torque M _{CORRECTION is calculated} according to the difference Δα _MODEL representing the time variation of the torsion. The PD member 5 corresponds to a PD controller known per se, wherein the ratios for the proportional and the differential part are adapted in experiments. In this case, a larger proportion of D acts stabilizing.

That of the PD member 5 Calculated correction torque M _CORRECTION becomes a predetermined by the driver torque M _{SOLL of} the internal combustion engine 1 in an adder 6 added. The result of this addition is the torque M ' _SOLL , which is the input signal for the actuator 2 and the predictor member 4 represents. Specifically, in this cycle, more and more improved torque specifications M ' _{SOLL can be} calculated by several iterative steps.

The damping device shown in particular suppresses torsional vibrations in the drive train very effectively 3 because it is not critical to stability as a control method due to dead times in the control loop. The internal combustion engine 1 namely has a dead time, which is mainly due to the burning process. The dead time of the internal combustion engine 1 is at a speed of 800 revolutions per minute (rpm) about 40 ms. The dead time is indirectly proportional to the speed. Because of this dead time is a measurement of the mechanical response of the powertrain 3 and the internal combustion engine 1 on the manipulated variable m of the adjusting device 2 only possible after this dead time.

In contrast, the predictor member 4 with the model of the powertrain 3 and the internal combustion engine 1 essentially no dead time. The time span after which the signal output of the predictor element 4 the response to the input M ' _{SOLL is} ready depends only on the computational speed of the predictor 4 from. The period of time is far less than the dead time of the internal combustion engine when using conventional microelectronic components 1 , Therefore, a timely calculation of a correction _torque M _{CORRECTION is} possible.

To check the predictive quality and to possibly model adapt the model of the predictor 4 comes with a measuring device 7 the actual time change Δα _IS the torsion of the drive train 3 measured. The measuring device 7 includes a speed sensor on the internal combustion engine 1 , which is the speed of the internal combustion engine 1 measures, and speed sensors on each driven wheel. Usually, in a motor vehicle, the rotational speeds of the internal combustion engine anyway 1 and the wheels measured, for example in the context of traction control. The measuring device 7 calculated from the signals of the individual speed sensors, the temporal change Δα _IST torsion of the drive train 3 , To this measured change in time Δα _IST the torsion of the drive train 3 With the calculated time change Δα _MODEL to be able to compare, it is necessary, the calculated state variable Δα _MODEL with a dead time element 8th to postpone. In a comparator unit 9 becomes the deadtime member 8th and the predictor member 4 calculated time change Δα ' _{MODEL of} the driveline torsion 3 with the measured time change Δα _IST the torsion of the drive train 3 compared. The result of this comparison represents the Error of prediction of the predictor member 4 The error serves as an input variable for an adaptation unit 10 , which has the task, the model of the predictor member 4 to adapt. This is done by parameter adjustment, for example, the spring and damping constants of the two-mass oscillator model. This ensures that the predictor member 4 even with changed mechanical properties of the internal combustion engine 1 and the powertrain 3 Continue correctly the answer of the powertrain 3 on a drive torque M ' _SOLL predicts.

In 2 schematically is the structure of a possible embodiment of the in the predictor member 4 contained in the LOLIMOT model. The illustrated LOLIMOT model consists of three local linear mapping functions (local linear models, LLM) 11.1 - 11.3 and the associated weighting functions 12.1 - 12.3 , At the entrance 13 is a signal, which contains the predetermined drive torque M ' _{target of} the internal combustion engine. The LOLIMOT model calculates the influence of the given drive torque M ' _SOLL on the oscillatory system of the drive train. The local linear mapping functions 11.1 - 11.3 each calculate a vector whose two components are the angular velocities α ₁ , α _{2 of} the internal combustion engine and the driven wheels. The LOLIMOT model uses the vector of angular velocities α ₁ , α ₂ as the input signal, since the predicted mechanical behavior of the drivetrain depends on the actual angular velocities α ₁ , α ₂ . In a difference element 14 the difference between the two components of the vector of the angular velocities α ₁ , α _{2 is} calculated. This difference corresponds to the time variation Δα _MODEL of the driveline torsion and is output at the output of the LOLIMOT model.

In the LOLIMOT model exist for certain areas of the state variables drive torque M ' _setpoint and rotational speed α ₁ , α _{2 of} the internal combustion engine and the wheels different local linear mapping functions 11.1 - 11.3 whose share of the result is determined by the weighting functions 12.1 - 12.3 is controlled. In multiplier elements 15.1 - 15.3 The result is one of the local linear mapping functions 11.1 - 11.3 consisting of a vector with two components with the scalar of the corresponding weighting function of the weighting functions 12.1 - 12.3 multiplied. This is the result of the local linear mapping function 11.1 in the multiplier element 15.1 with the corresponding weighting function 12.1 multiplied. The same applies to the local linear mapping functions 11.2 and 11.3 and the weighting function 12.2 . 12.3 with the multiplier elements 15.2 and 15.3 , Subsequently, the vectors of the weighted results in a sum member 16 summed. This sum then represents the result of the calculation of the LOLIMOT model, the result containing the rotational angular velocities α ₁ , α _{2 of} the internal combustion engine and the driven wheels.

To the Adapting the LOLIMOT model to the mechanical properties of the Drivetrain can the parameters of the weighting function, so for example their filter focus and their integral value changed become. Furthermore, you can the parameters of the local linear mapping function are changed. It is also possible, the LOLIMOT model further local linear mapping functions and weighting functions add. This can also be done in an automated way, as in the above-mentioned publication from 0. Nelles et al. is described.

In 3 is shown a modeling of the powertrain in the form of a two-mass torsional vibration model with spring / damper element. In one embodiment of the invention, this model is used to determine the mechanical response of the powertrain to a drive torque M ' _SOLL . The two-mass torsional vibration model consists of a smaller mass moment of inertia 17 , which represents the rotating parts of the internal combustion engine and has the value I ₁ . The moment of inertia 17 rotates at the angular velocity of the internal combustion engine α ' _1st Via a spring / damper element 18 is the first mass moment of inertia 17 with a second moment of inertia 19 connected. The spring / damper element 18 consists of a torsion spring with the spring constant k and a rotary damper with the damping constant c. The second moment of inertia 19 has the value I ₂ and represents the wheels and the mass of the vehicle, wherein the mass of the vehicle with a radius of gyration corresponding to the radius of the wheels in the calculation of the mass moment of inertia 19 received. Furthermore, in the calculation of the mass moment of inertia 19 a possible existing translation of the powertrain be considered. Furthermore, the angular velocity of the wheels α ' _{2 is} not the actual angular velocity of the wheels, but a multiple or a fraction of the actual angular velocity of the wheels corresponding to the transmission ratio of the drive train.

One possibility for calculating the response of the model of the drive train to a predefined drive torque M ' _SOLL is to numerically predict the rotational angular speeds of the internal combustion engine α' ₁ and the wheels α ' ₂ in a time step method. The difference between the two rotational angular velocities is then a measure of the torsion of the Powertrain.

The parameters of the model, in the illustrated case the first moment of inertia I ₁ , the second moment of inertia I ₂ , the torsion spring constant k and the damping constant c, can be predetermined or determined in the experiment. For example, the mass moment of inertia of the moving parts of the internal combustion engine is usually known. From the mass of the vehicle and the wheels, taking into account the transmission ratio of the drive train, the mass moment of inertia 19 be calculated. It should be noted that the second moment of inertia 19 depends on the load of the vehicle. Are the two mass moments of inertia 17 and 19 known, the torsion spring constant k and the rotational damper constant c can be calculated in the test and also on the moving vehicle with changes in the predetermined drive torque M ' _SOLL using the Differentialglei chung of the model. Like that in 2 LOLIMOT model shown also has this model as authoritative state variables, the predetermined drive torque M ' _SOLL , the rotational angular velocity of the internal combustion engine α' ₁ and the corrected by the gear ratio rotational angular velocity α ' _{2 of} the wheels.

The Invention is not limited to the embodiment described above and the method described limited but includes others Devices and methods, as far as they are of the inventive concept Make use.

Claims (21)

Damping device for the suppression of torsional vibrations in the drive train ( 3 ) an internal combustion engine ( 1 ), with - a detection device ( 4 . 7 ) for determining a torsion of the drive train ( 3 ) mechanical state variable (Δα _MODEL , Δα _ACT , α ₁ , α ₂ , α ' ₁ , α' ₂ ) and - an adjusting device ( 2 ) for controlling the internal combustion engine ( 1 ) with a manipulated variable (m) as a function of the determined state variable (Δα _MODEL , Δα _ACT , α ₁ , α ₂ , α ' ₁ , α' ₂ ), characterized in that the detection device ( 4 . 7 ) a predictor member ( 4 ), which is a model of the powertrain ( 3 ) and / or the internal combustion engine ( 1 ) and the state quantity (Δα _MODEL , α ₁ , α ₂ , α ' ₁ , α' ₂ ) in response to the drive train ( 3 ) and / or the internal combustion engine ( 1 ) to the manipulated variable (m) on the basis of the model, the model having adaptable model parameters. Damping device according to claim 1, characterized in that in the predictive element ( 4 ) model is essentially free of deadtime, whereas the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) and / or a measuring device ( 7 ) have a dead time (t _TOT ). Damping device according to claim 1 or 2, characterized in that an adaptation unit ( 10 ) is provided for adapting the model parameters while driving. Damping device according to one of the preceding claims, characterized in that the model is a multi-mass model, with which the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) is modeled with a vehicle mass. Damping device according to claim 4, characterized in that the multi-mass oscillator model is a 2-mass torsional oscillator model, the two mass moments of inertia ( 17 . 19 ), which by a spring / damper element ( 18 ), wherein the first mass moment of inertia ( 17 ) moving parts of the internal combustion engine ( 1 ), while the second mass moment of inertia ( 19 ) imitates the wheels with vehicle mass, whereas the spring / damper element ( 18 ) the drive train ( 3 ). Damping device according to claim 5, characterized in that the model parameters constants of the mass moment of inertia ( 17 . 19 ) and / or the spring / damper element ( 18 ) are. Damping device according to one of claims 1 to 3, characterized in that the model is a LOLIMOT model, which determines the mechanical behavior of the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) with vehicle mass with predetermined linear mapping functions ( 11.1 - 11.3 ). Damping device according to Claim 7, characterized in that the LOLIMOT model has weighting functions ( 12.1 - 12.3 ), which corresponds to the individual imaging functions ( 11.1 - 11.3 ) assigned. Damping device according to claim 8, characterized in that the model parameters parameters of the weighting functions ( 12.1 - 12.3 ) of the LOLIMOT model. Damping device according to one of claims 7 to 9, characterized in that the model parameters parameters of the linear mapping functions ( 11.1 - 11.3 ) of the LOLIMOT model. Motor control with a damping device after a of the preceding claims. Damping method for the suppression of torsional vibrations in the drive train ( 3 ) an internal combustion engine ( 1 ), comprising the following steps: - determining the torsion of the drive train ( 3 ) reproducing mechanical state variable (Δα _MODEL , Δα _actual , α ₁ , α ₂ , α ' ₁ , α' ₂ ) and - driving the internal combustion engine ( 1 ) with a manipulated variable (m) as a function of the determined state variable (Δα _MODEL , Δα _ACT , α ₁ , α ₂ , α ' ₁ , α' ₂ ), characterized by the following step: - determining the state variable (Δα _MODEL , α ₁ , α ₂ , α ' ₁ , α' ₂ ) in response to the manipulated variable (m) on the basis of a model of the drive train ( 3 ) and / or the internal combustion engine ( 1 ), the model having adaptable model parameters. Damping method according to claim 12, characterized in that the model is essentially free of deadtime, whereas the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) and / or a measuring device ( 7 ) have a dead time (t _TOT ). attenuation method according to claim 12 or 13, characterized in that the model parameters while adapted to the journey. Damping method according to one of claims 12 to 14, characterized in that with the model of the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) is modeled with a vehicle mass as a multi-mass model. Damping method according to claim 15, characterized in that the multi-mass oscillator model is a two-mass torsional oscillator model which has two mass moment of inertia ( 17 . 19 ), which are interconnected by a spring / damper element (18), wherein the first mass moment of inertia ( 17 ) moving parts of the internal combustion engine ( 1 ), while the second mass moment of inertia ( 19 ) imitates the wheels with vehicle mass, whereas the spring / damper element (18) the drive train ( 3 ). Damping method according to claim 16, characterized in that the model parameters are constants of the mass moment of inertia ( 17 . 19 ) and / or the spring / damper element (18). Damping method according to one of claims 12 to 14, characterized in that the model is a LOLIMOT model, which determines the mechanical behavior of the internal combustion engine ( 1 ) and / or the powertrain ( 3 ) with vehicle mass with predetermined linear mapping functions ( 11.1 - 11.3 ). Damping method according to claim 18, characterized in that the LOLIMOT model has weighting functions ( 12.1 - 12.3 ), which corresponds to the individual imaging functions ( 11.1 - 11.3 ) assigned. Damping method according to claim 19, characterized in that the model parameters are parameters of the weighting functions ( 12.1 - 12.3 ) of the LOLIMOT model. Damping method according to one of claims 18 to 20, characterized in that the model parameters parameters of the linear mapping functions ( 11.1 - 11.3 ) of the LOLIMOT model.